AZ-305 Practice Question 64: Time Series Telemetry Analytics
Microsoft Azure Solutions Architect Expert (AZ-305) · Test 7 · question 4 of 10
A manufacturer ingests about 2 TB a day of telemetry from 100,000 sensors. Engineers need interactive ad hoc queries over billions of rows that return in seconds, built-in time series functions for anomaly detection, and one year of retention. The team already writes queries in Kusto Query Language (KQL).
Which data store should you recommend?
Show answer and explanation
Answer:
Azure Data Explorer is a fast, fully managed analytics service built for large volumes of streaming log and telemetry data. It supports interactive exploration with KQL, includes a large set of time series analysis functions (such as anomaly detection), and is designed to scale to the ingest and query volumes described.
Option A is a general-purpose relational engine. It can hold large data, but it is not tuned for multi-terabyte-per-day telemetry ingestion and ad hoc scans, and it does not use KQL.
Option C is an operational document database. Its analytical store supports analytics on transactional data, but it is not designed as an interactive KQL engine for sensor telemetry at this scale.
Option D is a relational database whose extensions help with time series storage, but a single server is not designed to scan billions of rows interactively at 2 TB a day, and it does not use KQL.
Reference: What is Azure Data Explorer?